Acoustic Characterization of PSU Recording Studio Above and Below the Schroeder Frequency

Open Access
Kinzie, Andrew David
Graduate Program:
Master of Science
Document Type:
Master Thesis
Date of Defense:
August 30, 2017
Committee Members:
  • Daniel Allen Russell, Thesis Advisor
  • Michelle Celine Vigeant, Committee Member
  • David Carl Swanson, Committee Member
  • Acoustics
  • Recording
  • Studio
  • Room
  • Architecture
  • modes
  • Schroeder
  • Geometrical acoustics
  • Transmission Loss
Often times in architectural acoustics, there has been an emphasis on large rooms such as concert halls, auditoriums, and open atriums. This study conversely looks at the acoustics of small rooms with a focus on Penn State's recording Studio A in Music Building II. In all rooms, the full frequency response can be grouped into two distinct sections; separated by the Schroeder frequency ($f_s$). For many cases, rooms are large enough that most of the audible frequency range falls above the Schroeder frequency, where as smaller spaces have a significant audible range below. The goal of this study was to characterize the acoustic performance (in both ranges) of the control room and live room of Studio A and provide suggestions to reduce the impact of undesirable phenomena. Both rooms have important roles to play in the operation of a studio which depend on their acoustic character. COMSOL multi-physics models for each room were created to understand the modal behavior below the Schroeder frequency. For comparison, measured modes were visualized from frequency response measurement points in one plane at a 1 foot spacing. For frequencies above the Schroeder frequency, an ODEON model was created and validated against measured impulse responses, where metrics such as reverberation time (T60) and early decay time (EDT) have been considered. In addition, transmission loss measurements were conducted between the control room and live room. This allowed for analysis of the partition's performance for frequencies in one-third octave bands. By comparing the computer simulations to the measurements, it has been revealed that there are prevalent modes that occur in both rooms and match the simulated models reasonably well. Frequency response plots give a good indication of the Schroeder frequency and show where modal behavior begins and mid-to-high statistics take over. With reverberation times being quite short at around 0.2-0.3 seconds, relative changes in the model were made to give suggestions for increased performance. Improvements such as implementing the "live end, dead end" (LEDE) concept, and incorporating Helmholtz resonators have been explored and are intended to improve the acoustics and functionality of each space. In the event where acoustic issues are unavoidable, measurements will help aid audio engineers with insight on how they can adapt to the room and refine their craft.